Alzheimer's disease (AD) is a major health problem facing this country. The destruction of brain tissue in this degenerative disease likely begins decades before the onset of clinical symptoms. Identification of preclinical markers for AD would be extremely valuable for both intervention and treatment. Recent work suggests that decrements in cognition and changes on neuroimaging in asymptomatic individuals may identify those who go on to develop AD. The aim of this project is to study both cognition and structural and functional neuroimaging longitudinally in a sample of adults who are at increased risk for development of AD and contrast these findings with those of a matched control group. Specifically we will examine cognitive performance on tests of memory and learning, generalized and regional brain measures from conventional MRI and changes in activation with a memory fMRI paradigm. In a small subsample, we will investigate activation changes using olfactory fMRI. In addition, participants will be referred to NIH for assay of amyloid in cerebral spinal fluid. Participants, all at least 50 years of age, will include adult offspring (N=100) of autopsy-confirmed AD cases who are members of multiplex families with extensively characterized pedigrees, currently enrolled in a genetic study of AD, and adults matched for age and gender (N=100) who are presently followed in a study of normal aging. Approximately equal numbers of males and females will be enrolled. All participants will be evaluated twice, three years apart to examine change over time on these measures. In addition, all participants will be typed on two genetic markers (APOE, Alpha2 Macroglobulin) for correlation of allele status with the above measures. The results of this study will provide information on the identification of preclinical markers for late-onset AD. It is hoped that the findings here will prove of significant value as future interventions for Alzheimer's disease are developed. In addition, the proposed studies will provide valuable data for planned long-term longitudinal studies of persons at-risk for AD.